The invention relates to automatic power control in a data transmission system.
Cable modems allow computer systems to transmit and receive data over an ordinary cable television (TV) network. The cable modem receives a TV signal and downconverts this signal to obtain a digital signal suitable for computer operation. Because existing cable infrastructure does not provide a dedicated communication line for each subscriber, subscribers in an area must share a cable that leads to a server located at a head-end unit, known as a Cable Modem Termination System (CMTS). The CMTS serves all of the cable modems in a particular subscription area.
FIG. 1 shows the transmission path over a cable TV network that includes a head-end unit, or Cable Modem Termination System (CMTS) 102, and a cable modem 104 at a subscriber site. The head-end includes an interface 108 to a data-over-cable wide-area network (WAN) 100. Each subscriber site includes an interface 112 between the cable modem 104 and the subscriber computer 106. This interface 112, known as the cable-modem-to-customer-premises-equipment interface, usually includes a Peripheral Component Interface (PCI) bus, a Universal Serial Bus (USB), or an Ethernet type connection.
Signals transmitted from the cable modem 104 to the CMTS 102, i.e., in the xe2x80x9cup-streamxe2x80x9d direction, are sent in bursts so that many cable modems can transmit on the same frequency. This frequency sharing technique is commonly known as time-division multiple access (TDMA). In general, digital data from the subscriber computer 106 is encoded by the cable modem 104 on a burst-by-burst basis. A burst modulator in the cable modem 104 modulates the encoded signal using a common modulation technique. The modem 104 then transmits the signal to the CMTS 102 at an assigned carrier frequency, typically between 5 MHz and 42 MHz. Because this band is particularly susceptible to noise, including interference from Citizen Band (CB) radios and impulse noise from home appliances, most cable modems use a low order modulation scheme, such as QPSK (Quadrature Phase Shift Keying) or 16 QAM (Quadrature Amplitude Modulation), that is robust in noisy environments. In general, cable modems use more than one modulation technique, such as QPSK and 16 QAM, often switching among techniques from burst to burst.
Because the CMTS 102 must manage a variety of signals intended for multiple subscribers, the CMTS 102 sends control packets that provide the cable modem 104 at the subscriber site with instructions for sending and receiving data packets. A control packet usually indicates the frequency and the power level at which the cable modem 104 should transmit data packets, as well as the type of modulation to use for each data burst. The control packet also tells the cable modem 104 how many packets to transmit upstream and which downstream data packets are intended for the cable modem 104. From this information, the cable modem 104 determines FEC coding gains, signal-to-noise-ratio (SNR) requirements, packet lengths, and modulation types.
The RF communication path 110 and the structure for data and control packets sent over this communication path 110 are defined in the Data-Over-Cable Service Interface Specifications, Radio Frequency (RF) Interface Specification, SP-RFIv1.1-D01-98214, published by Cable Television Laboratories, Inc. This specification requires the expected average power level for upstream signals to remain constant. However, the average power level in a cable modem that uses multiple modulation techniques varies as the modulation technique changes from burst to burst. Therefore, cable modems that use multiple modulation techniques for upstream communication inherently violate the RF Interface Specification.
One aspect of the invention involves the transmission of digital data over a transmission channel, such as a cable TV network, by receiving blocks of data and modulating each block using one of at least two modulation types. A selected gain value is applied to each block of data before transmission. This gain value is selected from a group of at least two predetermined gain values, each of which corresponds to one of the modulation types.
In some embodiments, the gain values are selected and applied to the blocks of data by an automatic power control element that includes a gain element and a control element. The gain element applies the selected gain values to the blocks of data, and the control element selects the gain values from the group of predetermined values. In some cases, the control element includes a switch that selectively provides the gain values to the gain element in response to a signal indicating the modulation type to be used in transmitting each of the blocks of data.
In other embodiments, the modulation types have different associated average power levels. Suitable modulation types include QPSK and k-QAM, where k is a power of 2. In many of these embodiments, the gain values are selected to maintain substantially constant average transmission power level among all of the blocks of data. A digital storage device often is used to store the predetermined gain values.
In another aspect, the invention features an automatic power control system for use in transmitting digital data over a transmission channel, such as a cable TV network. The power control system includes a gain element that receives blocks of digital data, each of which is to be transmitted using one of at least two different modulation types. The gain element applies a selected gain value to each of the blocks of data before transmission. The power control system also includes a control element that selects the gain value to be applied to each of the blocks of data from a group of at least two predetermined gain values, each corresponding to one of the modulation types.
Among the advantages of the invention is the ability to maintain a substantially constant average power level from data burst to data burst, even when different modulation types are used. As a result, a data transmission system, such as a cable modem, can share a transmission channel and a power amplification circuit with other systems without cutting into the power allocated to the other systems.